The large-scale Quasar-Lyman α Forest Cross-Correlation from BOSS

TL;DR

This work measures the large-scale cross-correlation between quasars and the Lyα forest using ~60,000 DR9/BOSS quasar spectra, extending to separations up to $80\,h^{-1}\mathrm{Mpc}$. A linear-theory framework with quasar bias $b_q$, Lyα bias $b_F$, and redshift-space distortion parameters $\beta_q$, $\beta_F$ fits the data well for $r>15\,h^{-1}\mathrm{Mpc}$, yielding $b_q=3.64^{+0.13}_{-0.15}$ at $\bar{z}=2.38$ and $\beta_F=1.1\pm0.15$, in agreement with independent Lyα and quasar clustering measurements. The analysis finds evidence for a mean quasar redshift offset $\Delta_z\approx-160\,\mathrm{km\,s^{-1}}$ and a small redshift-error dispersion $\epsilon_z$, with the simple linear model failing at $r<15\,h^{-1}\mathrm{Mpc}$ likely due to quasar radiation effects. The paper discusses the implications of quasar radiation on the proximity effect and provides predictions for corrections in proximity measurements, while highlighting the potential to detect the BAO feature in the quasar-Lyα cross-correlation in future data.

Abstract

We measure the large-scale cross-correlation of quasars with the Lyman α forest absorption in redshift space, using ~ 60000 quasar spectra from Data Release 9 (DR9) of the Baryon Oscillation Spectroscopic Survey (BOSS). The cross-correlation is detected over a wide range of scales, up to comoving separations r of 80 Mpc/h. For r > 15 Mpc/h, we show that the cross-correlation is well fitted by the linear theory prediction for the mean overdensity around a quasar host halo in the standard ΛCDM model, with the redshift distortions indicative of gravitational evolution detected at high confidence. Using previous determinations of the Lyman α forest bias factor obtained from the Lyman α autocorrelation, we infer the quasar bias factor to be b_q = 3.64^+0.13_-0.15 at a mean redshift z=2.38, in agreement with previous measurements from the quasar auto-correlation. We also obtain a new estimate of the Lyman α forest redshift distortion factor, β_F = 1.1 +/- 0.15, slightly larger than but consistent with the previous measurement from the Lyman α forest autocorrelation. The simple linear model we use fails at separations r < 15 Mpc/h, and we show that this may reasonably be due to the enhanced ionization due to radiation from the quasars. We also provide the expected correction that the mass overdensity around the quasar implies for measurements of the ionizing radiation background from the line-of-sight proximity effect.

Figures (7)

• Figure 1: Left panel: Distribution of the 61342 quasar redshifts in our sample. Right panel: $i$-band absolute magnitude distribution.
• Figure 2: Measured cross-correlation in the indicated bins of perpendicular separation $\sigma$, as a function of the parallel separation $\pi$. The data points in green have a total separation $r=(\sigma^2 + \pi^2)^{1/2}<15 \, h^{-1} \, {\rm Mpc}$, and are not used in most of our fits. Solid (dashed) dark (blue) lines show the best fit model for the fiducial analysis, when using bins with separations down to $r=15 \, h^{-1} \, {\rm Mpc}$ ($r=7 \, h^{-1} \, {\rm Mpc}$).
• Figure 3: Two dimensional contours of the measured cross-correlation (left panel), compared to the best fit theoretical models for $r>15\, h^{-1} \, {\rm Mpc}$ (right panel). The black circle corresponds to $r=15 \, h^{-1} \, {\rm Mpc}$.
• Figure 4: Contours of $\chi^2$ in the two-parameter plane of the Ly$\alpha$ forest redshift distortion parameter versus: quasar bias (left), and quasar redshift error dispersion (right). The number of degrees of freedom is 130.
• Figure 5: Fitted QSO bias $b_q$ in several bins of the separation $r$, when fixing the other parameters to their best fit value in the fiducial model, $\beta_F=1.1$, $\epsilon_z=2.42 \, h^{-1} \, {\rm Mpc}$, $\Delta_z=-1.57 \, h^{-1} \, {\rm Mpc}$. The horizontal lines show the best fit and uncertainties when using all separations larger than $r>15\, h^{-1} \, {\rm Mpc}$, the scale that is marked with a vertical dotted line in the figure.